The issue is to avoid the storage and operational penalty. 100 x 100 matrix in 100 decimals vs 100 x 100 matrix in 50 decimals for many operations like copy, scale, etc. But accumulation of inner products, you want to avoid digit loss, e.g., A and B are long vectors -- say 100000 long, with a few "large" elements but most smaller. However 99000 "small" numbers could add up and ...
As with many such issues, it doesn't affect the vast majority of occurrences. Indeed few of my own. IEEE arithmetic usually does the job (I was one of the members of the original committee). This particular concern arose when trying to tease out a signal from a great deal of noise from a spacecraft sensor. JN On 2020-03-14 4:28 p.m., Jeff Newmiller wrote: > Not sure I understand the concern. IEEE 754 double precision floating point > was invented to allow for avoiding loss of precision when manipulating single > precision floating point numbers... but then C just ignores single precision > and you are expected to know that the precision of your answers may only be > relied on at single precision for certain operations. > > Wouldn't you just set Rmpfr precision to double your actual desired precision > and move on? Though I suppose you might consider more than doubling the > desired precision to deal with exponentiation [1]. > > [1] https://en.m.wikipedia.org/wiki/Extended_precision#Working_range > > On March 14, 2020 1:10:19 PM PDT, J C Nash <profjcn...@gmail.com> wrote: >> Rmpfr does "support" matrix algebra, but I have been trying for some >> time to determine if it computes "double" precision (i.e., double the >> set level of precision) inner products. I suspect that it does NOT, >> which is unfortunate. However, I would be happy to be wrong about >> this. >> >> JN >> >> On 2020-03-14 3:41 p.m., Bert Gunter wrote: >>> Read its documentation yourself and unless you have good reason not >> to, >>> always cc the list (which I have done here). >>> >>> >>> Bert Gunter >>> >>> "The trouble with having an open mind is that people keep coming >> along and >>> sticking things into it." >>> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) >>> >>> >>> On Sat, Mar 14, 2020 at 12:28 PM 林伟璐 <13917987...@163.com> wrote: >>> >>>> Thanks. Does it support matrix algebra? >>>> >>>> >>>> >>>> 林伟璐 >>>> 邮箱：13917987...@163.com >>>> >>>> >> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?iconUrl=http%3A%2F%2Fmail-online.nosdn.127.net%2F17c23bc1722125aa6261c33736f525c5.jpg&name=%E6%9E%97%E4%BC%9F%E7%92%90&uid=13917987541%40163.com&ftlId=1&items=%5B%22%E9%82%AE%E7%AE%B1%EF%BC%9A13917987541%40163.com%22%5D> >>>> >>>> 签名由 网易邮箱大师 <https://mail.163.com/dashi/dlpro.html?from=mail88> 定制 >>>> >>>> On 03/15/2020 03:18, Bert Gunter <bgunter.4...@gmail.com> wrote: >>>> Use google instead, as I recommended. If that's impossible in China, >> you >>>> should state that and what you tried (Baidu) in your query. >>>> >>>> You'll get the Rmpfr package. >>>> >>>> Bert Gunter >>>> >>>> "The trouble with having an open mind is that people keep coming >> along and >>>> sticking things into it." >>>> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) >>>> >>>> >>>> On Sat, Mar 14, 2020 at 11:48 AM 林伟璐 <13917987...@163.com> wrote: >>>> >>>>> Thanks, I tried Baidu but get nothing >>>>> >>>>> >>>>> >>>>> 林伟璐 >>>>> 邮箱：13917987...@163.com >>>>> >>>>> >> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?iconUrl=http%3A%2F%2Fmail-online.nosdn.127.net%2F17c23bc1722125aa6261c33736f525c5.jpg&name=%E6%9E%97%E4%BC%9F%E7%92%90&uid=13917987541%40163.com&ftlId=1&items=%5B%22%E9%82%AE%E7%AE%B1%EF%BC%9A13917987541%40163.com%22%5D> >>>>> >>>>> 签名由 网易邮箱大师 <https://mail.163.com/dashi/dlpro.html?from=mail88> 定制 >>>>> >>>>> On 03/15/2020 02:44, Bert Gunter <bgunter.4...@gmail.com> wrote: >>>>> Here's a novel idea: >>>>> Do a google search on "multiprecision computing package R" for an >> answer. >>>>> >>>>> Bert Gunter >>>>> >>>>> "The trouble with having an open mind is that people keep coming >> along >>>>> and sticking things into it." >>>>> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) >>>>> >>>>> >>>>> On Sat, Mar 14, 2020 at 10:36 AM 林伟璐 <13917987...@163.com> wrote: >>>>> >>>>>> Dear all >>>>>> >>>>>> >>>>>> I need a multiprecision computing package in R, if anyone in the >> list >>>>>> knows, please let me known... >>>>>> >>>>>> >>>>>> Many thanks >>>>>> >>>>>> >>>>>> Weilu Lin >>>>>> [[alternative HTML version deleted]] >>>>>> >>>>>> ______________________________________________ >>>>>> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see >>>>>> https://stat.ethz.ch/mailman/listinfo/r-help >>>>>> PLEASE do read the posting guide >>>>>> http://www.R-project.org/posting-guide.html >>>>>> and provide commented, minimal, self-contained, reproducible code. >>>>>> >>>>> >>> >>> [[alternative HTML version deleted]] >>> >>> ______________________________________________ >>> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see >>> https://stat.ethz.ch/mailman/listinfo/r-help >>> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >>> and provide commented, minimal, self-contained, reproducible code. >>> >> >> ______________________________________________ >> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. > ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
